Three broadly-neutralizing monoclonal antibodies (MAbs), b12, 2F5 and 2G12, have been cloned from the B cells of people with on-going HIV-1 infections. These MAbs target envelope proteins of HIV-1 and recent passive-immunization studies using them alone, and in combination, have shown protection of macaques against IV- and mucosal-challenge doses of pathogenic SHIV strains. These results indicate that broadly-neutralizing Abs can produce sterilizing protection against the virus. Our aim in designing an AIDS vaccine is to induce the same Ab specificities as b12, 2F5 and 2G12 by active immunization. We are developing peptides that bind tightly and specifically to these MAbs, and plan to use them in a novel immunization strategy to target the production of these same Ab specificities in naive animals. First, we will prime animals with envelope proteins to elicit small amounts of the targeted Abs in the polyclonal response against the whole protein. Next, the MAb-specific peptides will be used to boost the production of only the targeted Abs. Strong T-cell epitopes will be conserved between the Env protein primes and peptide boosts to maintain B-cell responses. MAbs b12, 2F5, and probably 2G12, have very long H3 hypervariable loops that "normal mice" cannot produce. Thus, we also must use animals that can produce such Abs. We have made the most progress in this approach with the b12 MAb. The peptide, B2.1, binds tightly and specifically to the b12 MAb. The structure of B2.1 in complex with b12 Fab has been elucidated, and is being used to further engineer the B2.1 peptide. Our published studies clearly show that the affinity of b12 is stronger for recombinant B2.1 fused to a larger protein than for its synthetic-peptide analog. Thus, we propose to immunize rabbits, XenoMouseTM animals, and eventually macaques, with phage bearing Env proteins, followed by boosts with phage displaying B2.1 peptide. Serum Ab titers will be tested for the presence of b12-like reactivity, and to understand the molecular and cellular bases of these responses, B-cells bearing anti-B2.1 Abs and anti-gp120 Abs will be isolated by FACS, and their expressed V-genes cloned using RT-PCR on single cells. Cloned Abs having the desired reactivities will be tested for HIV-1-neutralization. The assays should be sensitive enough to detect low levels of b12-like Abs and their B cells. In this way, high-titer responses may be built from initially low ones. We have developed peptide leads for Mabs 2F5 and 2G12, and are doing so for the newly-discovered Mabs 4E10 and Z13; these will be similarly tested.